Vision sensor, a key sensor for ADAS and autonomous driving was galloping ahead in the past year along with the relentless march of LiDAR and radar technologies.
Automotive camera is just tending to pack network clustering, night vision, inward-looking and 3D capabilities.
In 2018 Continental AG launched MFC500, its fifth-generation automotive camera with image resolution of 1-to-8 megapixels. Continental projects to mass-produce it in 2020. The camera is interlinked with the environment: by connecting it to the electronic horizon (eHorizon’) and Road Database’, road information and landmarks can be transmitted to and received from the cloud to locate the vehicle and enable proactive driving.
Bosch third-generation camera with a detection range of 150 meters will go into mass production in 2019. Bosch will also build an automotive camera perception network among its OEM customers worldwide through “Bosch Road Signature” (a crowd-sourced localization service that enables automated vehicles to determine their exact position).
What Bosch and Continental have done is identical with Mobileye’s autonomous driving roadmap.
But Mobileye has set about putting its technology into practical use when its peers are still working on development. Mobileye Aftermarket announced a raft of partnerships and collaborations with ride-hailing leaders, municipalities and governmental agencies that will enable the mapping of city streets around the world, including London and New York, through the deployment of Mobileye 8 Connect? in thousands of ride-hailing vehicles. From early 2019 on, the fleets equipped with Mobileye 8 Connect will harvest valuable information on city streets and infrastructure to create high-definition crowd-sourced maps through Mobileye’s Road Experience Management (REM) – a critical enabler of full autonomy.
Mobileye has partnered with BMW, Nissan and Volkswagen that collectively had two million cars with EyeQ4 system-on-chips collecting road data in 2018. In Japan, Mobileye collaborated with Zenrin and Nissan on using REM technology to collect data of all highways in the country. Mobileye and its Chinese partners like SAIC, Great Wall Motor and NIO work together to push on REM program.
There is reason to believe that Chinese IT giants will follow suit. Examples include Alibaba striving to develop cooperative vehicle infrastructure system (CVIS), Hikvision that enjoys superiorities in camera field and invests Wuhu Sensorthch Intelligent Technology Co., Ltd., a radar firm, and Tus-Holdings Co., Ltd which acquired Suzhou INVO Automotive Electronics Co., Ltd., a developer of ADAS. These players grab some first-mover advantage in vehicle-to-infrastructure (V2I) camera-based network clustering perception system.
Inward-looking: half of vision-based ADAS firms are developing DMS
Driver monitoring systems (DMS) is largely needed by commercial vehicle ADAS and will be a standard configuration for future L3 passenger cars. Half of vision-based ADAS companies are developing DMS, including EyeSight, Shenzhen Autocruis Technology Co., Ltd., Roadefend Vision Technology (Shanghai) Co., Ltd., Whetron Electronics Co., Ltd., Wuhan JIMU Intelligent Technology Co., Ltd., Minieye, Beijing Smarter Eye Technology Co., Ltd. and Black Sesame Technologies.
EyeSight, an Israeli start-up founded in 2005, provides driver monitoring, gesture recognition and user perception and analysis technologies. Its software-based automotive sensing solutions need support from infrared (IR) or time of flight (TOF) sensors.
Black Sesame Technologies has developed in-vehicle monitoring systems with capabilities of face recognition-based driver authority authentication, driver fatigue monitoring, bad driving behavior monitoring and occupant monitoring.
In February 2019 Shanghai Baolong Automotive Corporation released a range of automotive sensor products from dynamic vision sensor, 77G/24G radar, stereo forward-looking system, infrared thermal imager for night vision and diver warning system to face recognition system.
Continental AG rolled out MFC500, a fifth-generation automotive camera capable of night vision.
In 2018 Denso developed a new vision sensor which improves vehicle’s night vision by using a solid-state imaging device and unique lenses specifically designed for low-light conditions.
In 2018 Israel-based Foresight unveiled its QuadSight? quad-camera vision system. Using four-camera technology that combines two pairs of stereoscopic infrared and visible-light cameras, the system is designed to achieve near 100% obstacle detection with near zero false alerts in any weather or light conditions (including complete darkness, rain, haze, fog and glare).
3D sensing is now one of the key development orientations of camera technology.
AMS and Sunny Optical announced in November 2017 a joint plan to develop 3D sensing solutions for mobile devices and automotive application.
ZF is developing a kind of three-dimension interior observation system (3D IOS) that can detect occupants in a car and classify them to ascertain their sizes and positions as well as whether the occupant is manipulating the car willingly or monitoring the autonomous driving system.
Magna and Renesas collaborate in fusion of Magna’s eyeris? 3D surround-view camera system with the latter’s SoC (system-on-a-chip) to develop a more cost-effective 3D surround-view system especially for entry-level and mid-range cars.
Naturally, automotive 3D sensing not only relies on cameras but also takes advantage of either radar or LiDAR for 3D imaging. It is through using radar that Vayyar Imaging brings 3D imaging into a reality.
Vayyar Imaging announced in May 2018 to roll out millimeter wave 3D imaging system used chip. Vayyar’s 3D sensor can be utilized for obstacle detection, classification and simultaneous localization and mapping (SLAM). The embedded 3D sensor of Vayyar can scan the internal environment inside the car in real time and offer real-time imaging. Alerts will be sent to the drowsy driver through monitoring the vital signs of the occupant inside the car, or the guardian will be warned if the children and pets get off the car.
In 2018, the OEM installations of cameras (inclusive of front view, rear view, side view, interior view, surround view, driving recorder camera, among others.) to passenger car reached 20.584 million units with a year-on-year surge of 14.9%, of which front camera soared 28% YoY, surround-view camera shot up by 30.1% YoY, and stereo camera skyrocketed 170.3% YoY, according to ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Automotive Vision by ResearchInChina.
ADAS and Autonomous Driving Industry Chain Report, 2018-2019 of ResearchInChina covers following 17 reports:
1)Global Autonomous Driving Simulation and Virtual Test Industry Chain Report, 2018-2019
2)China Car Timeshare Rental and Autonomous Driving Report, 2018-2019
3)Report on Emerging Automakers in China, 2018-2019
4)Global and China HD Map Industry Report, 2018-2019
5)Global and China Automotive Domain Control Unit (DCU) Industry Report, 2018-2019
6)Global and China Automated Parking and Autonomous Parking Industry Report, 2018-2019
7)Cooperative Vehicle Infrastructure System (CVIS) and Vehicle to Everything (V2X) Industry Report, 2018-2019
8)Autonomous Driving High-precision Positioning Industry Report, 2018-2019
9)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Processor
10)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Automotive Lidar
11)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Automotive Radar
12)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Automotive Vision
13)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Passenger Car Makers
14)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- System Integrators
15)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Commercial Vehicle Automated Driving
16)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- Low-speed Autonomous Vehicle
17)ADAS and Autonomous Driving Industry Chain Report, 2018-2019- L4 Autonomous Driving
Follow us on Social Media: